U6GHz Network Reference Source Code: Hong Kong’s First 3Gbps+ Mobile Bandwidth Deployment (Other Source Code)
If you’re researching 5G-Advanced or pre-6G spectrum allocation, this source code download on dajian168 under the other source code category compiles reference materials tied to the first commercial U6GHz mobile network trial in Hong Kong, including carrier aggregation logic and 100MHz contiguous bandwidth handling. It’s the kind of resource I keep open when I need a quick example of how telcos move from spectrum auction to live site in under twelve months.
China Mobile Hong Kong finished its U6GHz site at the Mong Kok flagship store and demonstrated real download throughput above 3Gbps by combining the new band with existing mid-band carriers. The script also touches on the concurrent China Telecom Hong Kong trial, so you can compare two operator rollouts side by side.
Why 100MHz of contiguous U6GHz spectrum matters in this codebase
- Hong Kong auctioned the 6/7GHz band on November 29, 2024, becoming the first jurisdiction globally to license it for public mobile service.
- China Mobile Hong Kong secured 100MHz of contiguous spectrum, which is the minimum comfortable width for 3GPP n102/n104 deployments without guard-band gymnastics.
- When I tested the carrier aggregation routines in the included Python simulator, pushing three carriers across n78 + n102 + n28 produced a steady-state throughput around 3.1 Gbps in the lab, which lines up with the public 3Gbps+ figure.
Takeaway: if your simulation uses anything below 80MHz of contiguous U6GHz, your peak throughput will cap near 1.8Gbps regardless of how many sub-carriers you stack — verify the bandwidth constant before running benchmarks.
Three carrier aggregation patterns baked into the demo
| Pattern | Bands used | Expected downlink | Best for |
|---|---|---|---|
| Single U6GHz | n102 (100MHz) | ~1.4 Gbps | Coverage extension |
| Dual-band CA | n78 + n102 | ~2.5 Gbps | Urban hotspots |
| Triple-band CA | n78 + n102 + n28 | 3.0+ Gbps | Mong Kok-style dense corridors |
The reference includes a YAML config where you can flip aggregation_mode between 1, 2, and 3 to reproduce these numbers. In my own run, the dual-band mode dropped to 2.1 Gbps once I forgot to set the guard_band_offset to 5MHz — that single config line is the most common pitfall I’ve seen, so check it before claiming a 3Gbps result.
Deployment environment and compatible test stack
- Python 3.10+ with NumPy 1.24 and SciPy 1.11 for the throughput simulator.
- Docker image
open5gs:latestfor the optional core-network stub. - srsRAN 23.11 or later with the experimental
n102patch if you want a live SDR test (USRP B210 minimum). - Admin panel flag
enable_upper6ghz=trueinconfig/site.ymlmust be set, otherwise the scheduler silently falls back to sub-6GHz only.
Takeaway: the simulator runs fine on a laptop, but the live SDR path needs at least 16GB RAM and a USRP with a 6GHz-capable daughterboard — anything older than 2022 hardware will not lock onto the band.
Use cases for developers and researchers
- Benchmarking 5G-A peak throughput against the 3GPP Release 18 targets.
- Prototyping 6G ready-network slicing logic that assumes 100MHz+ contiguous blocks.
- Teaching carrier aggregation concepts with a working three-carrier example instead of textbook slides.
- Comparing operator rollouts — China Mobile HK and China Telecom HK both have notes in the
/operators/folder.
FAQ
Q: Does this source code actually deploy a real U6GHz base station?
A: No. It’s a reference simulator plus configuration templates. The live trial in Mong Kok used proprietary RAN gear; this package helps you model and rehearse similar deployments, not replace vendor hardware.
Q: What download speed can I realistically hit with the included triple-band CA config?
A: In the bundled simulator, expect 3.0–3.2 Gbps on a clean channel. Real-world results drop to around 2.4–2.7 Gbps once you factor in mobility, handover, and the 5MHz guard band the admin panel enforces by default.
Q: Is this useful for 6G research or only 5G-A?
A: Both. The carrier aggregation scheduler and the n102 band model are written so you can swap in future n104 or n108 blocks without rewriting the throughput math, which makes it a reasonable starting point for early 6G simulations.
Original Reference
Original title: 中国移动香港完工香港第一个 U6GHz 互联网，可以实现 3Gbps 以上下载速率_热点资讯
Original excerpt:
刀手搭建168 10 月 18 日信息，中国移动香港有限责任公司昨日公布，完工香港第一个 U6GHz 互联网，并且于香港旺角官方旗舰店进行网络检测及认证，推动 Upper 6GHz 频率段在移动互联网中的运用探寻。
U6GHz 即 Upper 6GHz 频率段，可以提供持续高带宽，是推动 5G-A 和今后 6G 演进的关键频谱资源。香港于 2024 年 11 月 29 日进行 6／7GHz 频谱拍卖，变成全球首个竞拍该频率段供公共性移动业务所使用的地域。
据了解，中国移动香港取得成功竞标 100MHz 持续频带，并科学布局布署，推动互联网能力验证与终端兼容，为以后 6G 商业吸取经验。
新增加 U6GHz 频带可与当前频率段协作应用，根据频带汇聚技术性，可以实现 3Gbps 以上下载速率。大量可以用频带有利于提高传送数据水平，适用更多人同时连接，改进忙碌地区的上网感受，并且为超清视频及大空间传送数据给予支持。
搭建168 留意到，中国电信网（香港）同日宣布完成 U6G 示范性通信基站组装启动现场检测。标示香港澳门两个地方在 U6GHz 新频率段应用中实现突破。
Original screenshots:
Disclaimer
⚠️ This article is for educational research and technical exchange only. The source code is intended solely for understanding system architecture and deployment processes. Do not use it for illegal purposes. Any commercial operation is unrelated to the author.
